# thresholds = {
# "Stress_initiation": (0, 99),
# "Aberrant_splicing": (1, 97),
# "Autophagy": (5, 97),
# "DAPI": (0.2, 95),
# "Apoptosis": (0.27, 95),
# "impaired_Autophagosome": (5, 99),
# "Cytoskeleton": (10, 99.8),
# "DNA_damage_P53BP1": (0.3, 90),
# "DNA_damage_pH2Ax": (7, 85),
# "mature_Autophagosome": (2, 98),
# "Necrosis": (4, 95),
# "Neuronal_activity": (5, 90),
# "Nuclear_speckles_SC35": (0.2, 85),
# "Nuclear_speckles_SON": (0.2, 85),
# "Parthanatos_early": (0, 85),
# "Parthanatos_late": (5, 95),
# "Protein_degradation": (5, 92),
# "Senescence_signaling": (10, 90),
# "Splicing_factories": (0, 95),
# "TDP-43": (5, 99.2),
# "Ubiquitin_levels": (2, 90),
# "UPR_ATF4": (2, 99.7),
# "UPR_ATF6": (3, 100),
# "UPR_IRE1a": (2, 98),
# "Necroptosis_pMLKL": (5, 90),
# "Necroptosis_HMGB1": (0.01, 90),
# }
# # Extended batch data with new entries
# data = [
# (2, 'DNA_damage_pH2Ax', 1, 1017118, 'Untreated'),
# (2, 'Parthanatos_early', 1, 1048087, 'stress'),
# (2, 'DNA_damage_pH2Ax', 2, 1008566, 'stress'),
# (2, 'Parthanatos_early', 1, 981344, 'stress'),
# (2, 'Nuclear_speckles_SON', 1, 981344, 'stress'),
# (1, 'Autophagy', 1, 1057052, 'stress'),
# (1, 'impaired_Autophagosome', 1, 1057052, 'stress'),
# (1, 'UPR_IRE1a', 1, 1057052, 'stress'),
# (1, 'Ubiquitin_levels', 1, 1057052, 'stress'),
# (1, 'UPR_IRE1a', 1, 981344, 'stress'),
# (1, 'Ubiquitin_levels', 1, 981344, 'stress'),
# (1, 'DNA_damage_P53BP1', 1, 981344, 'stress'),
# (1, 'Neuronal_activity', 1, 981344, 'stress'),
# (1, 'Necroptosis_HMGB1', 1, 981344, 'stress'),
# (1, 'Necrosis', 1, 981344, 'stress'),
# (1, 'Nuclear_speckles_SC35', 2, 981344, 'stress'),
# (1, 'Splicing_factories', 2, 981344, 'stress'),
# (1, 'Aberrant_splicing', 2, 981344, 'stress'),
# (1, 'Parthanatos_late', 2, 981344, 'stress'),
# (1, 'Senescence_signaling', 2, 1048087, 'stress'),
# (1, 'Apoptosis', 2, 981344, 'stress'),
# (1, 'Necroptosis_pMLKL', 2, 981344, 'stress'),
# (2, 'Autophagy', 1, 1017118, 'stress'),
# (2, 'impaired_Autophagosome', 1, 1017118, 'stress'),
# (2, 'UPR_IRE1a', 1, 1048087, 'Untreated'),
# (2, 'UPR_IRE1a', 1, 1008566, 'stress'),
# (2, 'Ubiquitin_levels', 1, 1048087, 'Untreated'),
# (2, 'Ubiquitin_levels', 1, 1008566, 'stress'),
# (2, 'DNA_damage_P53BP1', 1, 1017118, 'stress'),
# (2, 'Neuronal_activity', 1, 1017118, 'stress'),
# (2, 'DNA_damage_P53BP1', 2, 981344, 'stress'),
# (2, 'Neuronal_activity', 2, 981344, 'stress'),
# (2, 'DNA_damage_P53BP1', 1, 981344, 'stress'),
# (2, 'Neuronal_activity', 1, 981344, 'stress'),
# (2, 'DNA_damage_P53BP1', 1, 1057052, 'stress'),
# (2, 'Neuronal_activity', 1, 1057052, 'stress'),
# (2, 'DNA_damage_P53BP1', 1, 1005373, 'stress'),
# (2, 'Neuronal_activity', 1, 1005373, 'stress'),
# (2, 'DNA_damage_P53BP1', 2, 1057052, 'stress'),
# (2, 'Neuronal_activity', 2, 1057052, 'stress'),
# (2, 'Necroptosis_HMGB1', 1, 981344, 'Untreated'),
# (2, 'Necrosis', 1, 981344, 'Untreated'),
# (2, 'Necroptosis_HMGB1', 2, 981344, 'stress'),
# (2, 'Necrosis', 2, 981344, 'stress'),
# (2, 'Necroptosis_HMGB1', 2, 1001733, 'stress'),
# (2, 'Necrosis', 2, 1001733, 'stress'),
# (2, 'Necroptosis_HMGB1', 2, 1005373, 'stress'),
# (2, 'Necrosis', 2, 1005373, 'stress')
# ]
# # Cell line ID mapping
# cell_line_mapping = {
# 1008566: 'C9orf72-HRE_1008566',
# 981344: 'C9orf72-HRE_981344',
# 1001733: 'Control_1001733',
# 1017118: 'Control_1017118',
# 1025045: 'Control_1025045',
# 1048087: 'Control_1048087',
# 1057052: 'TDP--43-G348V_1057052',
# 1005373: 'TDP--43-N390D_1005373'
# }
# # Convert to DataFrame (renamed to `marker_df`)
# marker_df = pd.DataFrame(data, columns=['Batch', 'Name', 'Rep', 'CellLineID', 'Condition'])
# lower_thresh_DAPI, upper_thresh_DAPI = thresholds['DAPI']
# cb = get_specific_imgs(dfb, marker='DAPI')
# cb = cb.loc[(cb.Batch == 'Batch1') | (cb.Batch == 'Batch2')].Target_Sharpness_Brenner.values
# lower_p_DAPI = np.percentile(cb, lower_thresh_DAPI)
# upper_p_DAPI = np.percentile(cb, upper_thresh_DAPI)
# # Iterate over each row
# for _, row in marker_df.iterrows():
# batch, marker, rep, cell_line_id, condition = row
# # Match Cell Line Name
# cell_line = cell_line_mapping.get(cell_line_id, f'Unknown_{cell_line_id}')
# # Fetch images based on specific conditions
# cb1 = get_specific_imgs(dfb, marker=marker, batch=batch, rep=rep,
# cell_line=cell_line, condition=condition)
# dapi_paths = cb1.Path.apply(lambda p: get_dapi_path(p, marker1=marker, marker2='DAPI', funova=True))
# dfDAPI = dfb.loc[dfb['Path'].isin(dapi_paths)].Target_Sharpness_Brenner.values
# cb1 = cb1.Target_Sharpness_Brenner.values
# # Print statistics dynamically for the specific selection
# print(f"Stats for marker: {marker}, batch: {batch}, rep: {rep}, cell_line: {cell_line}, condition: {condition}")
# print(f"Mean: {np.mean(cb1):.2f}, Median: {np.median(cb1):.2f}, "
# f"10th Percentile: {np.percentile(cb1, 10):.2f}, "
# f"50th Percentile (Median): {np.percentile(cb1, 50):.2f}, "
# f"90th Percentile: {np.percentile(cb1, 90):.2f}")
# # Fetch images for the entire dataset with the given marker
# cb = get_specific_imgs(dfb, marker=marker)
# cb = cb.loc[(cb.Batch == 'Batch1') | (cb.Batch == 'Batch2')].Target_Sharpness_Brenner.values
# lower_thresh, upper_thresh = thresholds[marker]
# lower_p = np.percentile(cb, lower_thresh)
# upper_p = np.percentile(cb, upper_thresh)
# filtered_out = np.sum((cb1 < lower_p) | (cb1 > upper_p))
# total_count = len(cb1)
# filtered_percentage = (filtered_out / total_count) * 100 if total_count > 0 else 0
# filtered_out_DAPI = np.sum((dfDAPI < lower_p_DAPI) | (dfDAPI > upper_p_DAPI))
# total_count_DAPI = len(dfDAPI)
# filtered_percentage_DAPI = (filtered_out_DAPI / total_count_DAPI) * 100 if total_count_DAPI > 0 else 0
# # Print statistics dynamically for the full dataset
# print(f"Stats for marker: {marker} (All batches, reps, and cell lines)")
# print(f"Mean: {np.mean(cb):.2f}, Median: {np.median(cb):.2f}, "
# f"10th Percentile: {np.percentile(cb, 10):.2f}, "
# f"50th Percentile (Median): {np.percentile(cb, 50):.2f}, "
# f"90th Percentile: {np.percentile(cb, 90):.2f}")
# print(f"Threshold Percentiles ({lower_thresh}% - {upper_thresh}%): {lower_p:.2f} - {upper_p:.2f}")
# print(f"Filtered Out: {filtered_out}/{total_count} ({filtered_percentage:.2f}%)")
# print(f"Filtered Out_DAPI: {filtered_out_DAPI}/{total_count_DAPI} ({filtered_percentage_DAPI:.2f}%)")
# print("-" * 80) # Separator for readability
